Economic theory and psychology of non-use values.

I. INTRODUCTION

Both the Comprehensive Environmental Response, Compensation and Liability Act of 1980 (CERCLA, 42 USC 9601 et seq.) and the Oil Pollution Act of 1990 (OPA, 33 USC 2701 et seq.) provide a framework for the inclusion of non-use values in natural resource damage assessment. As the recent debate over the use of contingent valuation under OPA illustrates, many issues remain in non-market valuation which have significant public policy impacts. Similarly, as more Superfund sites enter the damage assessment phase following remediation under CERCLA, issues of natural resource damage assessment will reemerge. The appropriate inclusion of non-use values will be of primary concern.

In this paper we discuss the theoretical basis for use and non-use values for an environmental commodity. As discussed by Freeman (1993), the contingent valuation method (CVM) is the only available method for measuring non-use values. This paper builds on Freeman's discussion to further explore issues in the measurement of nonuse values using the contingent valuation method. We show that economic theory implies that the motives for and beliefs relating to non-use values must be carefully understood prior to summing values across individuals.

We use techniques drawn from psychology to explore these motives, which vary greatly across individuals, and develop a variety of theoretical models for the issue of cleaning up contaminated groundwater to explain observed responses consistent with these differing motives and beliefs. We show, both in theory and in our survey responses (McClelland et al. 1992; Lazo et al. 1992) that, depending on the individual's perception of the feasibility of intergenerational transfers (i.e., perfect versus imperfect water and/or capital markets) and the nature of an individual's preferences (i.e., paternalistic and/or nonpaternalistic altruism), stated willingness to pay for an environmental commodity may or may not include values that could be double-counted across generations. The possibility that the inclusion of bequest values can result in double counting (which depends on the assumption of nonpaternalistic altruism) is of particular concern to policymakers as these potentially constitute a significant portion of total values for programs such as cleanup of contaminated groundwater. The possibility of double counting has

led some opponents of the CVM to claim that all non-use values should be eliminated from benefit estimates. We show both theoretically and empirically that this conclusion is not warranted.

The methodology used for the development of the nationwide contingent valuation study of the benefits of groundwater cleanup lends insight into how such values can be interpreted for policy purposes. In pretesting the survey instrument we used verbal protocols, wherein subjects speak continuously into a tape recorder while completing the survey, as well as retrospective reports wherein subjects are individually debriefed immediately after completing the survey.(1) Examples of verbal protocol and retrospective report responses are given to motivate alternative theoretical models of non-use values. These lend support to the view that some individuals hold both paternalistic intergenerational preferences for groundwater cleanup and are motivated by perceived (inefficient) overuse of resources in providing substantial bequest values. Both paternalistic altruism and current overuse of a natural resource are shown theoretically to provide appropriate motives for bequest values which should be included in measures of environmental benefits.

The total value of groundwater cleanup or preservation can be defined as consisting of four components:

1. Use Value - the direct value to each household for the clean water they consume themselves (including any adjustment for uncertainty which has been termed option value);

2. Altruistic Value - the value that households place on other households having clean groundwater today;

3. Bequest Value - the value that the current generation places on the availability of clean groundwater to future generations;

4. Existence Value - the value that individuals place on simply knowing that groundwater is clean independent of any use, that is, the value for cleanup even if people never used the water.

The latter three categories are generally termed non-use values (see Krutilla 1967). The application of these value measures in the case of groundwater is not as straightforward as might be supposed. This occurs because of a possible confounding of use, altruistic, bequest, and existence values; because of the presence of paternalistic altruism; because water markets themselves are highly imperfect; and because capital markets are often perceived to be imperfect. In Section II we develop an intragenerational model to examine how nonpaternalistic altruism impacts stated values. In Section III, we extend this model to an intergenerational theoretical model consistent with most economic analysis (perfect markets and nonpaternalistic altruism) which likely obtains from some individuals. Then, motivated by our psychological study we modify this theoretical model in successive stages to account for the motives and beliefs expressed by some of our respondents.

II. A MODEL OF INTRAGENERATIONAL NONPATERNALISTIC ALTRUISM

To begin, we wish to consider the case of a society consisting of two individuals who share feelings of nonpaternalistic altruism.(2) That is, they both derive utility from each other's utility, not from the other's specific pattern of consumption across goods, public or private. Throughout the paper we use groundwater decontamination as our example of an environmental good since we obtained psychological insights for this commodity. Thus, these individuals, A and B, enjoy private consumption measured in dollars of [C.sup.A] and [C.sup.B], respectively, but derive disutility from knowing that contaminated groundwater in the amount Z is present. Thus, an existence value exists for clean groundwater as a pure public good derived from this dislike for contaminated ground-water. The utilities of our two individuals can then be written for A as

[U.sup.A] = [U.sub.A] (Z, [C.sup.A], [U.sup.B]) [1]

and for B as

[U.sup.B] = [U.sup.B](Z, [C.sup.B], [U.sup.A]) [2]

where [Mathematical Expression Omitted], [Mathematical Expression Omitted]; [Mathematical Expression Omitted], [Mathematical Expression Omitted], and [Mathematical Expression Omitted], [Mathematical Expression Omitted].(3) Where [Y.sup.A] and[Y.sup.B] are incomes and [T.sup.A] and [T.sup.B] are taxes collected to pay to decontaminate groundwater, the budget constraints are of the form for A.

[Y.sup.A] - [C.sup.A] - [T.sup.A] = 0, [3]

and for B,

[Y.sup.B] - [C.sup.B] - [T.sup.B] = 0. [4]

Where D is the amount of water decontaminated (a public good) and [Z.sup.0] is the initial amount of contaminated groundwater, the amount of contaminated groundwater "left" is

Z = [Z.sup.0] - D. [5]

Finally, taxes collected for decontamination must equal the cost of decontamination E(D) so we have

[T.sup.A] + [T.sup.B] - E(D) = 0 [6]

where E[prime](D), E[double prime](D) [greater than] 0.

The surprising result for nonpaternalistic altruism is that the condition for a Pareto optimal provision of the public good, D, is the same as the case where altruism is not present. The standard Samuelson conditions for the optimal provision of a public good hold where the sum of the marginal benefits equals the marginal cost of cleanup. This is obtained by maximizing [1] subject to a constrained level of B's utility [Mathematical Expression Omitted] and the other constraints [3]-[6] to …